| WARNING ⚠️ user config directory '/home/azureuser/.config/Ultralytics' is not writable, using '/tmp/Ultralytics'. Set YOLO_CONFIG_DIR to override. |
| Creating new Ultralytics Settings v0.0.6 file ✅ |
| View Ultralytics Settings with 'yolo settings' or at '/tmp/Ultralytics/settings.json' |
| Update Settings with 'yolo settings key=value', i.e. 'yolo settings runs_dir=path/to/dir'. For help see https: |
| CUDA initialization: The NVIDIA driver on your system is too old (found version 12080). Please update your GPU driver by downloading and installing a new version from the URL: http: |
| ====================================================================== |
| FINAL TRAINING ON H100 - BALANCED DATASET |
| ====================================================================== |
|
|
| GPU Available: False |
|
|
| ====================================================================== |
| STEP 1: Downloading Datasets from Roboflow |
| ====================================================================== |
|
|
| Dataset 1: New helmet images (212)... |
|
loading Roboflow workspace... |
|
loading Roboflow project... |
|
Downloading Dataset Version Zip in ~/helmet_212 to yolov8:: 0%| | 0/12500 [00:00<?, ?it/s]
Downloading Dataset Version Zip in ~/helmet_212 to yolov8:: 3%|▎ | 319/12500 [00:00<00:03, 3182.36it/s]
Downloading Dataset Version Zip in ~/helmet_212 to yolov8:: 21%|██ | 2625/12500 [00:00<00:00, 14777.55it/s]
Downloading Dataset Version Zip in ~/helmet_212 to yolov8:: 94%|█████████▍| 11768/12500 [00:00<00:00, 49648.18it/s]
Downloading Dataset Version Zip in ~/helmet_212 to yolov8:: 100%|██████████| 12500/12500 [00:00<00:00, 40786.43it/s] |
|
|
|
Extracting Dataset Version Zip to ~/helmet_212 in yolov8:: 0%| | 0/427 [00:00<?, ?it/s]
Extracting Dataset Version Zip to ~/helmet_212 in yolov8:: 100%|██████████| 427/427 [00:00<00:00, 12797.93it/s] |
|
|
| Dataset 2: No-helmet images (499)... |
|
loading Roboflow workspace... |
|
loading Roboflow project... |
|
Downloading Dataset Version Zip in ~/no_helmet_499 to yolov8:: 0%| | 0/32368 [00:00<?, ?it/s]
Downloading Dataset Version Zip in ~/no_helmet_499 to yolov8:: 1%| | 399/32368 [00:00<00:08, 3930.72it/s]
Downloading Dataset Version Zip in ~/no_helmet_499 to yolov8:: 20%|█▉ | 6318/32368 [00:00<00:00, 36234.63it/s]
Downloading Dataset Version Zip in ~/no_helmet_499 to yolov8:: 38%|███▊ | 12289/32368 [00:00<00:00, 38745.21it/s]
Downloading Dataset Version Zip in ~/no_helmet_499 to yolov8:: 50%|████▉ | 16132/32368 [00:00<00:00, 38523.52it/s]
Downloading Dataset Version Zip in ~/no_helmet_499 to yolov8:: 70%|██████▉ | 22529/32368 [00:00<00:00, 46709.73it/s]
Downloading Dataset Version Zip in ~/no_helmet_499 to yolov8:: 100%|██████████| 32368/32368 [00:00<00:00, 52238.32it/s] |
|
|
|
Extracting Dataset Version Zip to ~/no_helmet_499 in yolov8:: 0%| | 0/1003 [00:00<?, ?it/s]
Extracting Dataset Version Zip to ~/no_helmet_499 in yolov8:: 100%|██████████| 1003/1003 [00:00<00:00, 11989.05it/s] |
|
|
| Dataset 3: With-helmet images (300)... |
|
loading Roboflow workspace... |
|
loading Roboflow project... |
|
Downloading Dataset Version Zip in ~/with_helmet_300 to yolov8:: 0%| | 0/18485 [00:00<?, ?it/s]
Downloading Dataset Version Zip in ~/with_helmet_300 to yolov8:: 2%|▏ | 287/18485 [00:00<00:06, 2865.07it/s]
Downloading Dataset Version Zip in ~/with_helmet_300 to yolov8:: 24%|██▍ | 4425/18485 [00:00<00:00, 25503.78it/s]
Downloading Dataset Version Zip in ~/with_helmet_300 to yolov8:: 89%|████████▊ | 16385/18485 [00:00<00:00, 62736.87it/s]
Downloading Dataset Version Zip in ~/with_helmet_300 to yolov8:: 100%|██████████| 18485/18485 [00:00<00:00, 55136.84it/s] |
|
|
|
Extracting Dataset Version Zip to ~/with_helmet_300 in yolov8:: 0%| | 0/605 [00:00<?, ?it/s]
Extracting Dataset Version Zip to ~/with_helmet_300 in yolov8:: 100%|██████████| 605/605 [00:00<00:00, 13499.49it/s] |
|
|
| Dataset 4: Triple-riding (626)... |
|
loading Roboflow workspace... |
|
loading Roboflow project... |
|
Downloading Dataset Version Zip in ~/triple_riding_626 to yolov8:: 0%| | 0/30315 [00:00<?, ?it/s]
Downloading Dataset Version Zip in ~/triple_riding_626 to yolov8:: 1%| | 303/30315 [00:00<00:10, 2987.75it/s]
Downloading Dataset Version Zip in ~/triple_riding_626 to yolov8:: 9%|▊ | 2650/30315 [00:00<00:01, 14965.95it/s]
Downloading Dataset Version Zip in ~/triple_riding_626 to yolov8:: 39%|███▉ | 11837/30315 [00:00<00:00, 49968.55it/s]
Downloading Dataset Version Zip in ~/triple_riding_626 to yolov8:: 68%|██████▊ | 20481/30315 [00:00<00:00, 50538.10it/s]
Downloading Dataset Version Zip in ~/triple_riding_626 to yolov8:: 100%|██████████| 30315/30315 [00:00<00:00, 55162.02it/s] |
|
|
|
Extracting Dataset Version Zip to ~/triple_riding_626 in yolov8:: 0%| | 0/1264 [00:00<?, ?it/s]
Extracting Dataset Version Zip to ~/triple_riding_626 in yolov8:: 100%|██████████| 1264/1264 [00:00<00:00, 14533.21it/s] |
|
|
| ✅ All datasets downloaded! |
|
|
| ====================================================================== |
| STEP 2: Merging ALL Datasets |
| ====================================================================== |
|
|
| Unified classes (8): ['Helmet', 'Motorcycle', 'Rider', 'Triple Riding', 'helmet', 'more-than-2-person-on-2-wheeler', 'no helmet', 'with helmet'] |
|
|
| Copying datasets... |
| helmet212: 206 images |
| nohelmet499: 496 images |
| withhelmet300: 242 images |
| triple626: 626 images |
|
|
| Final merged dataset: |
| train: 1335 images |
| valid: 186 images |
| test: 116 images |
|
|
| Config saved: /home/azureuser/final_merged_h100/data.yaml |
|
|
| ====================================================================== |
| STEP 3: TRAINING ON H100 (96GB VRAM!) |
| ====================================================================== |
|
[KDownloading https:
[KDownloading https: |
|
|
| Training config: |
| Model: YOLO26m |
| Epochs: 150 (faster with H100) |
| Batch: -1 (auto - H100 can handle 64-128!) |
| Image size: 640 |
| Classes: 8 |
|
|
| Starting training... |
| Ultralytics 8.4.37 🚀 Python-3.12.3 torch-2.11.0+cu130 |
| Traceback (most recent call last): |
| File "/home/azureuser/train_h100_final.py", line 182, in <module> |
| results = model.train( |
| ^^^^^^^^^^^^ |
| File "/home/azureuser/yolo_h100_env/lib/python3.12/site-packages/ultralytics/engine/model.py", line 781, in train |
| self.trainer = (trainer or self._smart_load("trainer"))(overrides=args, _callbacks=self.callbacks) |
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ |
| File "/home/azureuser/yolo_h100_env/lib/python3.12/site-packages/ultralytics/models/yolo/detect/train.py", line 63, in __init__ |
| super().__init__(cfg, overrides, _callbacks) |
| File "/home/azureuser/yolo_h100_env/lib/python3.12/site-packages/ultralytics/engine/trainer.py", line 128, in __init__ |
| self.device = select_device(self.args.device) |
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ |
| File "/home/azureuser/yolo_h100_env/lib/python3.12/site-packages/ultralytics/utils/torch_utils.py", line 230, in select_device |
| raise ValueError( |
| ValueError: Invalid CUDA 'device=0' requested. Use 'device=cpu' or pass valid CUDA device(s) if available, i.e. 'device=0' or 'device=0,1,2,3' for Multi-GPU. |
|
|
| torch.cuda.is_available(): False |
| torch.cuda.device_count(): 1 |
| os.environ['CUDA_VISIBLE_DEVICES']: None |
|
|
|
|
